arXiv subjects

Ashish Gupta

Publications and source records attributed to Ashish Gupta.

At least 19 recordsLinked to original sources

arXiv2026

Bit-ViP: Leveraging Bit-planes to Preserve Visual Privacy in Images through Obfuscation

The unprecedented growth of computer vision applications, such as surveillance systems and social media, raises security and visual privacy concerns, especially when data is stored on cloud servers. Image obfuscation offers a way to preserve visual privacy while maintaining an adequate level of usability; thus, it has been a topic of great interest in recent years. However, prior obfuscation schemes are either vulnerable to malicious attacks, such as model inversion to reconstruct original images from obfuscated images, or generate non-trainable obfuscated images, making them unusable for achieving reasonable accuracy. This paper proposes a novel bit-plane-based image obfuscation scheme, {\em Bit-ViP}, to preserve visual privacy for image-based recognition tasks. The Bit-ViP scheme produces secure, usable images by incorporating an innovative end-to-end obfuscation function. While doing so, the obfuscated image would contain non-invertible noise (generated by Lorenz's chaotic system and differential privacy), making it hard for an adversary to reconstruct the original image. We conduct extensive experiments on two popular activity recognition datasets, namely UCF101 and HMDB51, to validate the effectiveness of Bit-ViP. In the face of attacks on reconstruction, pixel frequency, information entropy, and pixel inter-correlation, we present a rigorous security analysis demonstrating tangible improvements over existing schemes.

cs.CV↗

arXiv2026

Can Quantum Federated Learning Withstand Circuit-Level Backdoors?

Quantum Federated Learning (QFL) inherits the core vulnerability of federated optimization to malicious clients, while also introducing an attack surface from variational circuit training and measurement-driven gradients. This work proposes a novel CircUit-Level backdoor Threat (CULT) model that formalizes four stealthy attacks by exploiting quantum-aware mechanisms, including Grover, Pauli, Bit-flip, and Sign-flip. By enabling malicious clients on both in-training and post-training surfaces, these attacks can critically undermine the learning process. We establish a rigorous theoretical foundation to demonstrate attack stealthiness under standard smoothness assumptions. Experiments on the MNIST and CIFAR-10 datasets with non-IID splits and varying fractions of malicious clients show that even a single malicious client can induce severe accuracy degradation under FedAvg aggregation. While popular defenses, including Krum, Multi-Krum, FoolsGold, FLGuardian, and Mud-HoG, reduce degradation in many regimes, they fail to eliminate worst-case failure cases, where accuracy drops up to 50\%. The experimental analysis further reveals that under the CULT model, malicious updates effectively mask their presence by staying close to benign norms, thereby helping attackers evade detection.

quant-ph↗

arXiv2025

Energy-Efficient Split Learning for Resource-Constrained Environments: A Smart Farming Solution

Smart farming systems encounter significant challenges, including limited resources, the need for data privacy, and poor connectivity in rural areas. To address these issues, we present eEnergy-Split, an energy-efficient framework that utilizes split learning (SL) to enable collaborative model training without direct data sharing or heavy computation on edge devices. By distributing the model between edge devices and a central server, eEnergy-Split reduces on-device energy usage by up to 86 percent compared to federated learning (FL) while safeguarding data privacy. Moreover, SL improves classification accuracy by up to 6.2 percent over FL on ResNet-18 and by more modest amounts on GoogleNet and MobileNetV2. We propose an optimal edge deployment algorithm and a UAV trajectory planning strategy that solves the Traveling Salesman Problem (TSP) exactly to minimize flight cost and extend and maximize communication rounds. Comprehensive evaluations on agricultural pest datasets reveal that eEnergy-Split lowers UAV energy consumption compared to baseline methods and boosts overall accuracy by up to 17 percent. Notably, the energy efficiency of SL is shown to be model-dependent-yielding substantial savings in lightweight models like MobileNet, while communication and memory overheads may reduce efficiency gains in deeper networks. These results highlight the potential of combining SL with energy-aware design to deliver a scalable, privacy-preserving solution for resource-constrained smart farming environments.

cs.DC↗

arXiv2025

VideoMultiAgents: A Multi-Agent Framework for Video Question Answering

Video Question Answering (VQA) inherently relies on multimodal reasoning, integrating visual, temporal, and linguistic cues to achieve a deeper understanding of video content. However, many existing methods rely on feeding frame-level captions into a single model, making it difficult to adequately capture temporal and interactive contexts. To address this limitation, we introduce VideoMultiAgents, a framework that integrates specialized agents for vision, scene graph analysis, and text processing. It enhances video understanding leveraging complementary multimodal reasoning from independently operating agents. Our approach is also supplemented with a question-guided caption generation, which produces captions that highlight objects, actions, and temporal transitions directly relevant to a given query, thus improving the answer accuracy. Experimental results demonstrate that our method achieves state-of-the-art performance on Intent-QA (79.0%, +6.2% over previous SOTA), EgoSchema subset (75.4%, +3.4%), and NExT-QA (79.6%, +0.4%). The source code is available at https://github.com/PanasonicConnect/VideoMultiAgents.

cs.CV↗

arXiv2025

When Federated Learning Meets Quantum Computing: Survey and Research Opportunities

Quantum Federated Learning (QFL) is an emerging field that harnesses advances in Quantum Computing (QC) to improve the scalability and efficiency of decentralized Federated Learning (FL) models. This paper provides a systematic and comprehensive survey of the emerging problems and solutions when FL meets QC, from research protocol to a novel taxonomy, particularly focusing on both quantum and federated limitations, such as their architectures, Noisy Intermediate Scale Quantum (NISQ) devices, and privacy preservation, so on. With the introduction of two novel metrics, qubit utilization efficiency and quantum model training strategy, we present a thorough analysis of the current status of the QFL research. This work explores key developments and integration strategies, along with the impact of QC on FL, keeping a sharp focus on hybrid quantum-classical approaches. The paper offers an in-depth understanding of how the strengths of QC, such as gradient hiding, state entanglement, quantum key distribution, quantum security, and quantum-enhanced differential privacy, have been integrated into FL to ensure the privacy of participants in an enhanced, fast, and secure framework. Finally, this study proposes potential future directions to address the identified research gaps and challenges, aiming to inspire faster and more secure QFL models for practical use.

cs.DC↗

arXiv2025

Bernstein-type inequalities for quantum algebras

We establish Bernstein-type inequalities for the quantum algebras $K_{n,\Gamma}^{P,Q}(\mathbb{K})$ introduced by K. L. Horton that include the graded quantum Weyl algebra, the quantum symplectic space, the quantum Euclidean space, and quantum Heisenberg algebra etc., obtaining new results and as well as simplified proofs of previously known results. The Krull and global dimensions of certain further localizations of $K_{n,\Gamma}^{P,Q}(\mathbb{K})$ are computed.

math.QA↗

arXiv2024

Tackling Selfish Clients in Federated Learning

Federated Learning (FL) is a distributed machine learning paradigm facilitating participants to collaboratively train a model without revealing their local data. However, when FL is deployed into the wild, some intelligent clients can deliberately deviate from the standard training process to make the global model inclined toward their local model, thereby prioritizing their local data distribution. We refer to this novel category of misbehaving clients as selfish. In this paper, we propose a Robust aggregation strategy for FL server to mitigate the effect of Selfishness (in short RFL-Self). RFL-Self incorporates an innovative method to recover (or estimate) the true updates of selfish clients from the received ones, leveraging robust statistics (median of norms) of the updates at every round. By including the recovered updates in aggregation, our strategy offers strong robustness against selfishness. Our experimental results, obtained on MNIST and CIFAR-10 datasets, demonstrate that just 2% of clients behaving selfishly can decrease the accuracy by up to 36%, and RFL-Self can mitigate that effect without degrading the global model performance.

cs.LG↗

arXiv2024

Linear Algebra and Galois Theory

In \cite{GQ2008} R. Gow and R. Quinlan have cast a new look on the endomorphism algebra of a $K$-vector space $V$ of dimension $n$ assuming that $K$ has a Galois extension $L$ of degree $n$. In this approach the $K$-space $L$ may serve as a model for $V$ and Galois-theoretic ideas and results may be applied to elucidate the structure of endomorphisms and other important objects of linear algebra. In particular, this leads to the clarification of the structure of a rank-one endomorphism, trace of an endomorphism, criteria for linear indepedence etc. We present an exposition of these results using the language of tensor algebra wherever possible to provide shorter and more conceptual proofs.

math.RT↗

arXiv2024

NurtureNet: A Multi-task Video-based Approach for Newborn Anthropometry

Malnutrition among newborns is a top public health concern in developing countries. Identification and subsequent growth monitoring are key to successful interventions. However, this is challenging in rural communities where health systems tend to be inaccessible and under-equipped, with poor adherence to protocol. Our goal is to equip health workers and public health systems with a solution for contactless newborn anthropometry in the community. We propose NurtureNet, a multi-task model that fuses visual information (a video taken with a low-cost smartphone) with tabular inputs to regress multiple anthropometry estimates including weight, length, head circumference, and chest circumference. We show that visual proxy tasks of segmentation and keypoint prediction further improve performance. We establish the efficacy of the model through several experiments and achieve a relative error of 3.9% and mean absolute error of 114.3 g for weight estimation. Model compression to 15 MB also allows offline deployment to low-cost smartphones.

cs.CV↗

arXiv2023

Smart Connected Farms and Networked Farmers to Tackle Climate Challenges Impacting Agricultural Production

To meet the grand challenges of agricultural production including climate change impacts on crop production, a tight integration of social science, technology and agriculture experts including farmers are needed. There are rapid advances in information and communication technology, precision agriculture and data analytics, which are creating a fertile field for the creation of smart connected farms (SCF) and networked farmers. A network and coordinated farmer network provides unique advantages to farmers to enhance farm production and profitability, while tackling adverse climate events. The aim of this article is to provide a comprehensive overview of the state of the art in SCF including the advances in engineering, computer sciences, data sciences, social sciences and economics including data privacy, sharing and technology adoption.

cs.CY↗

arXiv2023

Towards a debuggable kernel design

This paper describes what it means for a kernel to be debuggable and proposes a kernel design with debuggability in mind. We evaluate the proposed kernel design by comparing the iterations required in cyclic debugging for different classes of bugs in a vanilla monolithic kernel to a variant enhanced with our design rules for debuggability. We discuss the trade offs involved in designing a debuggable kernel.

cs.OS↗

arXiv2023

Constant rank subspaces of alternating bilinear forms from Galois Theory

Let $L/K$ be a cyclic extension of degree $n = 2m$. It is known that the space $\text{Alt}_K(L)$ of alternating $K$-bilinear forms (skew-forms) on $L$ decomposes into a direct sum of $K$-subspaces $A^{\sigma^i}$ indexed by the elements of $\text{Gal}(L/K) = \langle \sigma \rangle$. It is also known that the components $A^{\sigma^i}$ can have nice constant-rank properties. We enhance and enrich these constant-rank results and show that the component $A^\sigma$ often decomposes directly into a sum of constant rank subspaces, that is, subspaces all of whose non-zero skew-forms have a fixed rank $r$. In particular, this is always true when $-1 \not \in L^2$. As a result we deduce a decomposition of $\text{Alt}_K(L)$ into subspaces of constant rank in several interesting situations. We also establish that a subspace of dimension $\frac{n}{2}$ all of whose nonzero skew-forms are non-degenerate can always be found in $A^{\sigma^i}$ where $\sigma^i$ has order divisible by $2$.

math.RA↗

arXiv2023

Triviality of the automorphism group of the multiparameter quantum affine $n$-space

A multiparameter quantum affine space of rank $n$ is the $\mathbb F$-algebra generated by indeterminates $X_1, \cdots, X_n$ satisfying $X_iX_j = q_{ij} X_jX_i \ (1 \le i < j \le n)$ where $q_{ij}$ are nonzero scalars in $\mathbb F^\ast$. The corresponding quantum torus is generated by the $X_i$ and together with their inverses subject to the same relations. So far the automorphisms of a quantum affine space have been considered mainly in the uniparameter case, that is, $q_{ij} = q$. We remove this restriction here. Necessary and sufficient conditions are obtained for the quantum affine space to be rigid, that is, the only automorphisms are the trivial ones arising from the action of the torus $(\mathbb F^\ast)^n$. These conditions are based on the multiparameters $q_{ij}$ and also on the subgroup of $\mathbb F^\ast$ generated by these multiparameters. We employ the results in J. Alev and M. Chamarie, Derivations et automorphismes de quelques algebras quantiques, Communications in Algebra, 1992 (20), 1787-1802, and point out a small error in a main theorem in this paper which however remains valid with a small modification. We also note that a quantum affine space whose corresponding quantum torus has dimension one necessarily has a trivial automorphism group. This is a consequence of a result of J.~M.~Osborne, D.~S.~Passman, Derivations of Skew Polynomial Rings, J. Algebra, 1995, 176, 417--448. We expand the known list of examples of quantum tori that have dimension one and are thus hereditary noetherian domains.

math.RA↗

arXiv2023

Calculation of true coincidence summing correction factor for clover detector in add-back and direct mode

The true coincidence summing effect on the full-energy peak efficiency calibration of an unsuppressed clover HPGe detector has been studied. Standard multi-energetic and mono-energetic gamma-ray sources were used to determine the full-energy peak efficiency of the detector as a function of the gamma-ray energies at different source-to-detector distances. The true coincidence summing correction factors for the full-energy peak efficiency of the detector has been determined, in the add-back and direct modes of the detector, using both experimental and analytical methods. Geant4 simulations were performed to obtain the full-energy peak efficiency and total efficiency of the detector for different gamma-ray energies. The simulated efficiencies were used to calculate the correction factors using the analytical method. The correction factors obtained from both analytical and experimental methods were found to be in good agreement with each other. The clover detector in add-back mode exhibits larger summing corrections compared to the direct mode for the same source-to-detector distances. For the add-back mode, the coincidence summing effect is not significant for source-to-detector distances ~ 13 cm or above, whereas, for the direct mode, measurements can be performed for source-to-detector distances ~ 5 cm or above without considering the coincidence summing effect.

physics.ins-det↗

arXiv2023

Is Meta-Learning the Right Approach for the Cold-Start Problem in Recommender Systems?

Recommender systems have become fundamental building blocks of modern online products and services, and have a substantial impact on user experience. In the past few years, deep learning methods have attracted a lot of research, and are now heavily used in modern real-world recommender systems. Nevertheless, dealing with recommendations in the cold-start setting, e.g., when a user has done limited interactions in the system, is a problem that remains far from solved. Meta-learning techniques, and in particular optimization-based meta-learning, have recently become the most popular approaches in the academic research literature for tackling the cold-start problem in deep learning models for recommender systems. However, current meta-learning approaches are not practical for real-world recommender systems, which have billions of users and items, and strict latency requirements. In this paper we show that it is possible to obtaining similar, or higher, performance on commonly used benchmarks for the cold-start problem without using meta-learning techniques. In more detail, we show that, when tuned correctly, standard and widely adopted deep learning models perform just as well as newer meta-learning models. We further show that an extremely simple modular approach using common representation learning techniques, can perform comparably to meta-learning techniques specifically designed for the cold-start setting while being much more easily deployable in real-world applications.

cs.IR↗

arXiv2023

Physics-informed radial basis network (PIRBN): A local approximating neural network for solving nonlinear PDEs

Our recent intensive study has found that physics-informed neural networks (PINN) tend to be local approximators after training. This observation leads to this novel physics-informed radial basis network (PIRBN), which can maintain the local property throughout the entire training process. Compared to deep neural networks, a PIRBN comprises of only one hidden layer and a radial basis "activation" function. Under appropriate conditions, we demonstrated that the training of PIRBNs using gradient descendent methods can converge to Gaussian processes. Besides, we studied the training dynamics of PIRBN via the neural tangent kernel (NTK) theory. In addition, comprehensive investigations regarding the initialisation strategies of PIRBN were conducted. Based on numerical examples, PIRBN has been demonstrated to be more effective and efficient than PINN in solving PDEs with high-frequency features and ill-posed computational domains. Moreover, the existing PINN numerical techniques, such as adaptive learning, decomposition and different types of loss functions, are applicable to PIRBN. The programs that can regenerate all numerical results can be found at https://github.com/JinshuaiBai/PIRBN.

cs.LG↗

arXiv2023

Calculation of true coincidence summing correction factor for a Broad Energy Germanium (BEGe) detector using standard and fabricated sources

The true coincidence summing (TCS) correction factor for a Broad Energy Germanium (BEGe) detector has been calculated at far and close geometry measurement using multi-energetic radioactive $\gamma$-ray sources $^{60}$Co, $^{133}$Ba and $^{152}$Eu. The correction factors were calculated using experimental method and analytical method. Photopeak efficiency and total efficiency required to calculate the correction factor were obtained using Geant4 Monte Carlo simulation code. A few standard as well as fabricated mono-energetic sources were also included in the $\gamma$-ray efficiency measurements. The simulated efficiencies of mono-energetic $\gamma$-ray sources were matched to experimental $\gamma$-ray efficiencies by optimizing the detector parameters. The same parameters were used to obtain the photopeak and total efficiency for $\gamma$-ray of our interest and coincident $\gamma$-ray. Analytical correction factors and experimental correction factors were found in good agreement with each other.

physics.ins-det↗

arXiv2022

Systematic investigation of channel coupling effects on elastic, inelastic and neutron transfer channels in $^6$Li+$^{159}$Tb

Elastic scattering angular distribution for weakly bound nucleus $^{6}$Li on the deformed rare earth $^{159}$Tb target nucleus has been measured at energies around the Coulomb barrier. The elastic scattering cross sections for this reaction consist of inelastic contributions from low lying excited states of $^{159}$Tb. The pure elastic cross-sections have been extracted from the admixture of elastic and inelastic data. The optical model potential parameters for the system have been obtained from the extracted pure elastic scattering cross sections. Coupled channel calculations have been performed with this set of potential parameters, to compare the theoretical and experimental inelastic scattering cross sections. The work has been extended to obtain the spectroscopic factor for $^{158}$Tb+n configuration from the experimental 1n-pickup data.

nucl-ex↗